Personality and E-petition Success: Perspectives of Online Leadership

2023 | Americas Conference on Information Systems | Citations: 0

Authors: Wu, Jiao; Wang, Ying

Abstract: Online petitions have been a popular and effective means for the public to voic ... Expand

Abstract: Online petitions have been a popular and effective means for the public to voice their opinions and potentially influence society. E-petitioners can be regarded as influencers or leaders who leverage their social impacts beyond the digital world. In the current study, we collected 41,952 e-petitions from change.org and used an advanced text-mining model (Roberta-base model) to detect petitioners' Big Five personality traits. Then, by applying the trait theory and leadership literature, we developed a research model and provided an in-deep understanding of how petitioners' personalities can influence the petition's success. Our findings indicate that while neuroticism turns out to have a positive effect, agreeableness and conscientiousness are significant negative factors that can hinder the success of an online petition. Collapse

Topics: personality government system

Methods: regression analysis method deep learning natural language processing survey

Theories: big five model leadership theory personality theory

The Effects of Quote Retweet on Subsequent Posting Behavior and Morality Expression on Social Media

2023 | HICSS | Citations: 0

Authors: Wu, Yan; Du, Qianzhou; Zhang, Xiaohui; Zhang, John

Abstract: Guided by the Threshold model and Self-justification theory, we propose and test ... Expand

Abstract: Guided by the Threshold model and Self-justification theory, we propose and test a research model regarding the impact of online discussion activity on users’ behaviors on social media. Specifically, we examine the effects of quote retweeting a tweet related to immigration policies and border issues on users’ subsequent posting behaviors and morality expression. In addition, we test the moderating effect of individual threshold level and behavior-opinion inconsistency on the main effect. Results indicate that individuals, who quote retweeted the selected-topic tweets, are likely to post more topic-related tweets and express more on morality. This impact can be strengthened when individuals have higher threshold levels or larger behavior-opinion inconsistency. These findings provide both theoretical and practical implications for social media governance. Collapse

Topics: Twitter social media climate change pandemic criminality

Methods: experimental group propensity score matching natural language processing econometric modeling deep learning

Theories: self justification theory cognitive dissonance theory

How the Uniqueness of Initial Posts Affect User Participation in Online Cancer Communities

2023 | Americas Conference on Information Systems | Citations: 0

Authors: Xu, Anqi; Gao, Yuanyuan

Abstract: Online cancer communities (OCCs) are crucial resources for cancer patients, pro ... Expand

Abstract: Online cancer communities (OCCs) are crucial resources for cancer patients, providing support for the complex and challenging aspects of the disease. Patients often face unique problems, and OCCs offer a place to seek support when friends and family may not understand. This study explores how unique topics discussed in OCCs impact user participation in a thread, measured by the number of replies, repliers, length of replies, and response speed. A deep-learning-based natural language processing topic model is used to calculate the uniqueness scores of posts in OCCs, based on the prevalence of each topic. The study has implications for OCC administrators, researchers, and physicians in terms of understanding and addressing unique health topics expressed in online communities. Collapse

Topics: participatory design database system online community

Methods: design methodology topic model econometric modeling natural language processing deep learning

A twin data-driven approach for user-experience based design innovation

2023 | International Journal of Information Management | Citations: 0

Authors: Yang, Bai; Liu, Ying; Chen, Wei

Abstract: Data-driven innovation has received increasing attention, which explores big dat ... Expand

Abstract: Data-driven innovation has received increasing attention, which explores big data technologies to gain more insights and advantages for product design. In user experience (UX) based design innovation, user-generated data and archived design documents are two valuable resources for various design activities such as identifying op­ portunities and generating design ideas. However, these two resources are usually isolated in different systems. Additionally, design information typically represented based on functional aspects is limited for UX-oriented design. To facilitate experience-oriented design activities, we propose a twin data-driven approach to inte­ grate UX data and archived design documents. In particular, we aim to extract UX concepts from product reviews and design concepts from patents respectively and to discover associations between the extracted concepts. First, a UX-integrated design information representation model is proposed to associate capabilities with key elements of UX at the concept, category, and aspect levels of information. Based on this model, a twin data-driven approach is developed to bridge experience information and design information. It contains three steps: expe­ rience aspect identification using an attention-based LSTM (Long short-term memory) network, design infor­ mation categorization based on topic clustering using BERT (Bidirectional Encoder Representations from Transformers) and LAD (Latent Dirichlet allocation) model, and experience needs and design information inte­ gration by leveraging word embedding techniques to measure concept similarity. A case study using healthcarerelated experience and design information has demonstrated the feasibility and effectiveness of this approach. Collapse

Topics: innovation management online review user experience big data idea generation

Methods: BERT case study machine learning natural language processing product review

What Makes AI Different? Exploring Affordances and Constraints - The Case of Auditing

2023 | European Conference On Information Systems | Citations: 0

Authors: Yang, Jiaqi; Marrone, Mauricio; Amrollahi, Alireza

Abstract: This study aims to gain a comprehensive understanding of the differences betwee ... Expand

Abstract: This study aims to gain a comprehensive understanding of the differences between classic IT and AI artefacts. To achieve this objective, the study employs a grounded theory literature review approach and analyses 81 papers related to the application of classic IT and AI artefacts in the auditing industry. Drawing on the Technology Affordances and Constraints Theory, we examine the actions that can be potentially enabled or restricted by using classic IT and AI artefacts. This analysis allows us to conceptualise and compare the affordances and constraints associated with these two types of artefacts. The study addresses the need for more research on AI from both social and technical perspectives. Our findings may facilitate practitioners in improving their business processes and promoting effective collaboration between humans and AI. Collapse

Topics: artificial intelligence affordance auditing information system artificial general intelligence weak AI

Methods: machine learning literature study image processing business process modeling design artifact

Theories: theory of affordance

Getting Personal: A Deep Learning Artifact for Text-Based Measurement of Personality

2023 | Information Systems Research | Citations: 0

Authors: Yang, Kai; Lau, Raymond Y. K.; Abbasi, Ahmed

Abstract: Analysts, managers, and policymakers are interested in predictive analytics capa ... Expand

Abstract: Analysts, managers, and policymakers are interested in predictive analytics capable of offering better foresight. It is generally accepted that in forecasting scenarios involving organizational policies or consumer decision making, personal characteristics, including personality, may be an important predictor of downstream outcomes. The inclusion of personality features in forecasting models has been hindered by the fact that traditional measurement mechanisms are often infeasible. Text-based personality detection has garnered attention because of the public availability of digital textual traces. However, the text machine learning space has bifurcated into two branches: feature-based methods relying on manually crafted human intuition, or deep learning language models that leverage big data and compute, the main commonality being that neither branch generates accurate personality assessments, thereby making personality measures infeasible for downstream forecasting applications. In this study, we propose DeepPerson, a design artifact for text-based personality detection that bridges these two branches by leveraging concepts from relevant psycholinguistic theories in conjunction with advanced deep learning strategies. DeepPerson incorporates novel transfer learning and hierarchical attention network methods that use psychological concepts and data augmentation in conjunction with person-level linguistic information. We evaluate the utility of the proposed artifact using an extensive design evaluation on three personality data sets in comparison with state-of-the-art methods proposed in academia and industry. DeepPerson can improve detection of personality dimensions by 10–20 percentage points relative to the best comparison methods. Using case studies in the finance and health domains, we show that more accurate text-based personality detection can translate into significant improvements in downstream applications such as forecasting future firm performance or predicting pandemic infection rates. Our findings have important implications for research at the intersection of design and data science, and practical implications for managers focused on enabling, producing, or consuming predictive analytics.History: Olivia Sheng, Senior Editor; Huimin Zhao, Associate Editor.Funding: This work was supported by Oracle for Research (NLP for the Greater Good) and the U.S. National Science Foundation’s Division of Information and Intelligent Systems [Grants BDS-1636933 and IIS-1816504]. The work was also partly supported by the Research Grants Council of the Hong Kong Special Administrative Region [Project: CityU 11507219] and the City University of Hong Kong SRG [Project: 7005780].Supplemental Material: The online appendices are available at https://doi.org/10.1287/isre.2022.1111. Collapse

Topics: personality Twitter social media pandemic Google+

Methods: machine learning BERT design science deep learning computational algorithm

Theories: big five model

Responding to Online Reviews in Competitive Markets: A Controlled Diffusion Approach

2023 | Management Information Systems Quarterly | Citations: 0

Authors: Yang, Mingwen; Zheng, Zhiqiang (Eric); Mookerjee, Vijay; Chen, Hongyu

Abstract: We study how firms respond to online customer reviews in a competitive market w ... Expand

Abstract: We study how firms respond to online customer reviews in a competitive market where they jostle with one another for sales based on online ratings. The focus of this paper is on how firms can optimally manage their ratings through management response and how review ratings affect the sales and profits of competing firms. We develop a controlled diffusion process to model the coevolution of sales and ratings as a function of the response strategy chosen to maximize profit over time. Our model considers a variety of factors, such as profit margin and customer rating sensitivity, that influence a firm's effort to manage ratings and subsequently its sales and profits. More response effort needs to be exerted to manage ratings when either the profit margin of a tour is very high or customers are very sensitive to ratings. We estimate our model using data on Ctrip's tours that include each tour's sales, reviews, prices, and tour features. We find that consumers anchor their beliefs in the mean market rating and that their purchase decisions depend on the tour's rating relative to this anchor. Thus, relative, rather than absolute, ratings matter. Our study informs firms on how competition and other primitives impact their efforts to manage ratings and hence profit. Our methodology allowed us to conduct "what-if" analyses, for example, to study what would happen to the review ratings, sales, and profits of a tour if a firm adopted a different response strategy. We were also able to provide turnaround strategies for struggling tours, i.e., factors that a loss-making tour should change if it wishes to make a positive profit. Ultimately, we conducted a competitive analysis that allowed us to modify certain parameters that affect the intensity of competition and hence the sales and the profits of competing tours. Finally, we demonstrate the flexibility of the model by extending it to incorporate multiple state variables that might affect the response strategy. Collapse

Topics: application programming interface online learning system Twitter innovation management service quality

Methods: longitudinal research time series analysis generalized autoregressive conditional heteroskedasticity model natural language processing autoregressive moving average model

Theories: co-evolution theory

sDTM: A Supervised Bayesian Deep Topic Model for Text Analytics

2023 | Information Systems Research | Citations: 0

Authors: Yang, Yi; Zhang, Kunpeng; Fan, Yangyang

Abstract: Topic modeling methods such as latent Dirichlet allocation (LDA) are powerful to ... Expand

Abstract: Topic modeling methods such as latent Dirichlet allocation (LDA) are powerful tools for analyzing massive amounts of textual data. They have been used extensively in information systems (IS) and business discipline research to identify latent topics for data exploration and as a feature engineering mechanism to derive new variables for analyses. However, existing topic modeling approaches are mostly unsupervised and only leverage textual data, while ignoring additional useful metadata often associated with text, such as star ratings in customer reviews or categories of posts in online forums. As a result, the identified topics and variables derived based on the learned topic model may not be accurate, which could lead to incorrect estimations that affect subsequent empirical analysis and to inferior performance on predictive tasks. In this study, we propose a novel supervised deep topic modeling approach called sDTM, which combines a neural variational autoencoder model and a recurrent neural network. sDTM leverages the auxiliary data associated with text to enhance the topic modeling capability. We conduct empirical case studies and predictive analytics on an online consumer review data set and an online knowledge community data set. Experimental results show that in comparison with benchmark methods, sDTM can enhance both the empirical estimation and predictive performance. sDTM makes methodological contributions to the IS literature and has direct relevance for research using text analytics.History: Alok Gupta, Senior Editor; Huimin Zhao, Associate Editor.Supplemental Material: The online appendix is available at https://doi.org/10.1287/isre.2022.1124. Collapse

Topics: online review open source user-generated content Wikipedia accounting

Methods: topic model deep learning machine learning BERT data transformation

Unifying Algorithmic and Theoretical Perspectives: Emotions in Online Reviews and Sales

2023 | Management Information Systems Quarterly | Citations: 2

Authors: Yu, Yifan; Yang, Yang; Huang, Jinghua; Tan, Yong

Abstract: Emotion artificial intelligence, the algorithm that recognizes and interprets v ... Expand

Abstract: Emotion artificial intelligence, the algorithm that recognizes and interprets various human emotions beyond valence (positive and negative polarity), is still in its infancy but has attracted attention from industry and academia. Based on discrete emotion theory and statistical language modeling, this work proposes an algorithm to enable automatic domain-adaptive emotion lexicon construction and multidimensional emotion detection in texts. Using a large-scale dataset of China's movie market from 2012 to 2018, we constructed and validated a domain-specific emotion lexicon and demonstrated the predictive power of eight discrete emotions (i.e., surprise, joy, anticipation, love, anxiety, sadness, anger, and disgust) in online reviews on box office sales. We found that representing overall emotions through discrete emotions yields higher prediction accuracy than valence or latent emotion variables generated by topic modeling. To understand the source of the predictive power from a theoretical perspective and to test the cross-culture generalizability of our prediction study, we further conducted an experiment in the U.S. movie market based on theories on emotion, judgment, and decision-making. We found that discrete emotions, mediated by perceived processing fluency, significantly affect the perceived review helpfulness, which further influences purchase intention. Our work shows the economic value of emotions in online reviews, generates insight into the mechanism of their effects, and has managerial implications for online review platform design, movie marketing, and cinema operations. Collapse

Topics: online review Python user-generated content microblog marketing management

Methods: computational algorithm word2vec parametric test product review experiment

Theories: psychological theory

A deep learning approach for detecting fake reviewers: Exploiting reviewing behavior and textual information

2023 | Decision Support Systems | Citations: 0

Authors: Zhang, Dong; Li, Wenwen; Niu, Baozhuang; Wu, Chong

Abstract: Ensuring the credibility of online consumer reviews (OCRs) is a growing societal ... Expand

Abstract: Ensuring the credibility of online consumer reviews (OCRs) is a growing societal concern. However, the problem of fake reviewers on online platforms significantly influences e-commerce authenticity and consumer trust. Existing studies for fake reviewer detection mainly focus on deriving novel behavioral and linguistic features. These features require extensive human labor and expertise, placing a heavy burden on platforms. Therefore, we propose a novel end-to-end framework to detect fake reviewers based on behavior and textual information. It has two key components: (1) a behavior-sensitive feature extractor that learns the underlying patterns of reviewing behavior; (2) a context-aware attention mechanism that extracts valuable features from online reviews. We rigorously evaluate each proposed module and the entire framework against state-of-the-art benchmarks on two real-world datasets from http://Yelp.com. Experimental results demonstrate that our method achieves state-ofthe-art results on fake reviewer detection. Our method can be considered a tentative step toward lowering human labor costs in realizing automated fake reviewer detection on e-commerce platforms. Collapse

Topics: fake review online review electronic commerce deception detection shapley additive explanation

Methods: machine learning digital trace data experiment parametric test deep learning